Managerial Economics II Study Guide

Carlos Cevallos

AEM 2601: Managerial Economics Study Guide for Prelim #1 - March 7, 2013

STATISTICS

§Managers need statistics (evidence) in order to make informed decisions, including:

oInformation on profits and sales, sales staff performance tied to compensation, firms considering buying vs. financing, etc.

§Example from class: scalping at a Lady Gaga concert

§"Statistical inference" way of thinking about a sample drawn from a population of interest

§Variables can be either quantitative or categorical

§Each visual representation of data (distribution, scatterplot, etc.) needs:

oUnits, labels, title(s), interval spacing validity/uniformity

§Quantitative data analysis includes:

oMin and Max (range), mean, median, mode, and distribution

§Symmetry and mean are the most important characteristics to look for in a distribution

o68%, 95%, 99.7%

omean<median=skewed left (presence of outliers)

omean=median=symmetric

omean>median=skewed right (presence of outliers)

§When interpreting data, ask:

oHow is that measured? What is the sample?

o Relative to what? What does "big" mean?

o What assumptions are being made and how do they affect the results?

oIs causation implied when in fact it's correlation?

§Consider omitted variables, sample selection bias, and simultaneous causality

STATISTICS/CAUSALITY

§The response variable is the outcome Y-variable, whereas the explanatory variable is the independent X-variable which explains why the responsible variable changes

§When interpreting relationships, consider:

oForm: linear, curved, clustered, no pattern (random)

oDirection: positive, negative, or no direction (random)

oStrength: how closely the points fit the "form"

oOutliers: deviations from the pattern with a low probability of occurrence

§Regression analysis: = + , wherein:

ois the intercept, the value of y when x is zero

ois the slope, the amount y changes when x increases by one unit

ois the predicted value of y given values for a and b

§Example from class: you work in marketing at shoeworld.com analyzing your product lines and you are trying to figure out the relationship between price charged and quantity demanded on eBay; Y=- 2X+60 à therefore, every additional dollar charged is associated with a 2 unit decrease in demand

§Why relationships are not necessarily explained by causality:

oOmitted variables are other variables that are related to your explanatory variable and that determine, at least in part, the response variable

oSample selection bias occurs when the availability of data is influenced by a selection process that is related to the value of the dependent variable

oSimultaneous causality is when in addition to a casual link from X to Y, there exists a casual link from Y to X

§Example from class: the most famous case of simultaneous causality is the "chicken or the egg"

§Absolutely everything must be held constant, otherwise changes in Y may not be attributable to a change in X

§In reality, we don't live in an ideal world, so instead:

oLook for the best counterfactual, that is, a comparison unit that did not receive "treatment" or estimating what would have happened to Y in the absence of treatment

Carlos Cevallos

AEM 2601: Managerial Economics Study Guide for Prelim #1 - March 7, 2013

oRandomized control experiments employ methods that explicitly measure casual impact, including the random assignment of treatment to units - treatment that is independent of any other determinants of outcome (thereby eliminating the possibility of omitted variable bias)

§Example from class: randomized control experiment in which the opening bid for shoes on eBay was lowered and shipping costs increased, leading to earlier and higher bids

STRATEGY

§Strategists/managers are interested in the reasons behind heterogeneity in firm performance

§Sun Tzu's The Art of War was the first popular book on strategy

o"Every battle is won before it is ever fought" à research, research, research

§Among strategy's objectives:

oGive direction and purpose

oDeploy resources in the most effective manner

oCoordinate decisions

§No firm "is an island," rather, it exists in an ecosystem composed of competitors, stakeholders, customers, as well as the economy-at-large

§Defining strategy

oThings change, and firms must plan accordingly - or preferably, in anticipation. Adaptation is

a necessary (but not sufficient) condition for survival

oCorporate strategy entails where a firm exists in the corporate environment

...